Fast deep reinforcement learning using online adjustments from the past.
Steven HansenAlexander PritzelPablo SprechmannAndré BarretoCharles BlundellPublished in: NeurIPS (2018)
Keyphrases
- reinforcement learning
- online learning
- real time
- model free
- learning algorithm
- neural network
- function approximation
- decision trees
- supervised learning
- search space
- state space
- search algorithm
- learning problems
- markov decision processes
- least squares
- expert systems
- social networks
- artificial intelligence
- machine learning
- real world
- data sets